Algorithm::Evolutionary::Run 0.77

Algorithm::Evolutionary::Run is a Perl class for setting up an experiment with algorithms and population.

SYNOPSIS

 use Algorithm::Evolutionary::Run;

 my $algorithm = new Algorithm::Evolutionary::Run 'conf.yaml';
 #or
 my $conf = {
 'fitness' => {
 'class' => 'MMDP'
 },
 'crossover' => {
 'priority' => '3',
 'points' => '2'
 },
 'max_generations' => '1000',
 'mutation' => {
 'priority' => '2',
 'rate' => '0.1'
 },
 'length' => '120',
 'max_fitness' => '20',
 'pop_size' => '1024',
 'selection_rate' => '0.1'
 };

 my $algorithm = new Algorithm::Evolutionary::Run $conf;

 #Run it to the end
 $algorithm->run();
 
 #Print results
 $algorithm->results();
 
 #A single step
 $algorithm->step();

This is a no-fuss class to have everything needed to run an algorithm in a single place, although for the time being it's reduced to fitness functions in the A::E::F namespace, and binary strings. Mostly for demo purposes, but can be an example of class for other stuff.


Product's homepage

Requirements:

· Perl

What's New in This Release: [ read full changelog ]

· lib/Algorithm/Evolutionary.pm: Added tests to see what can make some of them fail